Purpose: We compared two series of total hip arthroplasties (THA) at more than 12 years. The only difference between the two series was the nature of the femoral head: 28-mm zincronia head versus 32-mm alumina head.

Material and method: All THAs were performed during the same period (1988–1990) and the polyethylene inserts, the femoral components, the surgical cement and the operator were the same for both series. Patient age and mean weight were the same for the two series of 40 Zir/PE and 62 AL/PE implants: 98 and 63 years and 72 and 73 kg respectively. Implant survival, polyethylene wear on AP and lateral x-rays, and Merckel osteolysis were studied.

Results: Three revisions were required at eight, ten and eleven years in the Zir/PE group. No revisions were required in the Al/PE group where there were no signs of loosening on the radiograms.

During the first four years, polyethylene wear appeared to be the same in the two series. Beyond the fifth year, penetration of the alumina heads exhibited a regular mean 0.07 mm progression per year. Beyond the fifth year, penetration of the zircona heads accelerated reaching a mean 0.4 mm per year at twelve years. Mean volumetric wear of the polyethylene cups was 1360 mm3 for the zircona heads and 755 mm3 for the alumina heads. Osteolysis was seen as a defected facing the merckel and measured 2.5 cm2 for the Zir/PE couple and 0.35 cm2 for the Al/PE couple. The differences between the two series in linear penetration, volumetric wear, and osteolysis were significant (p < 0.05). The three zircona heads that were removed showed partial transformation from the tetragonal phase to the monoclinical phase (19%, 25% and 30%) and a roughened surface which was not sufficient to explain the polyethylene wear. Abnormal wear was explained more by loss of spherical shape and the greater volume of the zircona heads, probably related to a modification of the crystalline phase (change from the tetragonal to the monoclinical phase usually is associated with a 3% increase in volume). Analysis of the fingerprints left on the morse code of the zircona heads suggest the change in volume of the zircona heads was associated with a change in the contact between the morse cone and the zircona head over time. Analysis of the explanted cups showed wear with delamination in two cases. For one of the implants, the polyethylene insert was deformed exhibiting a fusion aspect corresponding to abnormal increase in temperature.

Discussion and conclusion: The zircona implants manufactured more than twelve years ago probably do not meet current manufacturing standards. Nevertheless, the changes observed in the explanted pieces would suggest they are related not only to the manufacturing process but also to the material itself.